Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 683
    Prev
    Next
    INDEX
    Explanations

    adverbs expressing a high degree of a quality or action

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    478
    +0.24
    0.8%
    468
    +0.14
    0.5%
    605
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.24
    0.04
    1776
    +0.14
    0.03
    468
    +0.14
    0.03
    Negative Logits
     abstrait
    -1.01
     keramik
    -0.97
     réaliste
    -0.95
     kontinu
    -0.93
     karton
    -0.91
     kosme
    -0.89
     silikon
    -0.89
     mikrofon
    -0.89
     kompres
    -0.89
     étrange
    -0.88
    POSITIVE LOGITS
     extremely
    0.63
     incredibly
    0.61
    Molto
    0.60
     zelo
    0.57
     felici
    0.57
    TargetApi
    0.55
    lably
    0.54
    pertise
    0.54
     velkommen
    0.54
    ecutive
    0.53
    Activations Density 0.109%

    No Known Activations